The slow pace of antibiotic discovery is being outmatched by rapid acquisition of resistance by our pathogens. The need is especially acute in the case of M. tuberculosis, where strains of XDR-TB resistant to the majority of standard therapeutics are rapidly spreading. In this project, we will combine two innovations to develop a drug discovery platform for identifying leads that will be subsequently developed into therapeutics for treating tuberculosis: the use of uncultured bacteria as a unique source of novel antimicrobials;and a high-throughput screen for specific anti-Mtb compounds. Most antibiotics in use today are natural products or their derivatives, obtained from screening soil microorganisms. The main practical problem with discovery from culturable microorganisms is the enormous background of known compounds. At the same time, the vast majority of bacteria, 99% of species, do not readily grow in vitro and are known as uncultured. Our group developed a general method to grow uncultured bacteria by cultivating them in situ. An environmental sample such as soil is mixed with agar and sandwiched between two semi-permeable membranes of a diffusion chamber which is returned to the environment. Isolated colonies of diverse organisms grow in the chamber, and subsequent reinoculation to new chambers produces "domesticated" variants capable of growing on regular Petri dishes in vitro. Our preliminary findings show that this is an excellent source of novel antimicrobials. However, even with this previously inaccessible resource most of the chemistry effort is still wasted on rediscovery of known compounds. We reason that the problem can be resolved if discovery is focused on a species-specific compound. In case of M. tuberculosis, several synthetic compounds have been discovered that specifically act against this organism - INH, ethionamide and pyrazinamide. Natural compounds specifically acting against M. tuberculosis have not been described so far. This means that a screen for specific anti-M. tuberculosis compounds will produce hits that will have a high probability of being novel substances. A specific screen will then largely replace the laborious dereplication. The rationale is to screen in parallel against M. tuberculosis and a different organism, S. aureus. A pilot screen showed an excellent specific hit rate of 1.5% for extracts from uncultured species acting against M. tuberculosis. This screen will be optimized in the proposed project. In order to properly validate the screen, we will need to demonstrate that it is indeed capable of identifying novel compounds acting specifically against M. tuberculosis. We will therefore dereplicate the hits, the structure of unknowns will be determined, and their mode of action will be established. Once developed and validated in this project, the screen will be used in an HTS format for large-scale drug discovery. A combination of a unique, untapped source - uncultured bacteria - and a specific screen is likely to lead to novel compounds to combat drug-resistant M. tuberculosis. PUBLIC HEALTH RELEVANCE: In this project, we will establish a method that allows to rapidly identify antimicrobial compounds for developing drugs to treat tuberculosis. The method is based on using a unique source of antimicrobial compounds - bacteria that do not normally grow in the lab and are known as "unculturable". We will also look for compounds that act specifically against the pathogen M. tuberculosis, which will avoid killing of the good bacteria of our gut flora.